Leak Testing Large Castings
Testing engine blocks, transmission cases, oil pans, aluminum wheels and other large automotive castings can be challenging. For instance, a bare engine block can be more than 2 feet long and weigh more than 85 pounds.
Leak testing big parts is not as easy as it is for smaller components, such as fluid reservoirs, air filters or gas caps. Cycle time and accuracy are challenges, mainly due to the size of the part under test. Sealing is also a challenge, because of the surface texture associated with cast parts.
Fortunately, equipment suppliers have developed some innovative ways for engineers to get the job done. Most automotive leak testing applications use automation because of high volumes and throughput rates. But, weight becomes an important factor with bigger auto parts.
“Large castings are almost 100 percent tested using automation, due to weight constraints of the test part, as well as potential safety concerns,” says Chris Goebel, senior director of component sales and marketing at ULVAC Technologies Inc. “The majority of large castings use roller conveyors to move parts throughout the factory.”
“Typically the challenge is not the test, but rather part handling in and out of the test fixture,” adds Rod Sano, marketing coordinator at VIC Leak Detection. “The larger the test part, in terms of dimensions and weight, the more challenging it is to load and unload.”
Material handling is often done manually during design and development, then fully automated for production. “We typically see manual load and unload operations with bigger parts,” notes Jacques Hoffman, president of InterTech Development Co. “Some manufacturers use robotics to automate the process. But, ergonomics should always be considered. For instance, test fixtures should be easily accessible and easy to reach.”
When leak testing large castings, engineers should be aware of upstream processes that can change the temperature of a part and affect test results. “Upstream processes that affect the temperature, like welding, can be problematic for pressure decay testing, which relies on a stable temperature,” warns John McLaren, marketing manager for leak detection in the Vacuum Products Division of Agilent Technologies Inc.
Engineers should always be on the lookout for potential contaminants, such as hydrocarbons used during the casting process. “Cutting and grinding fluids, and other residual oils, can contribute to outgassing and contaminate the test chamber,” claims Margaret Bishop, business development manager at Abbess Instruments. “This may result in false results and cross-contamination of other products. The larger the surface area, and the greater porosity and number of cavities in the part, there’s more risk of outgassing.”
“Large parts can also be warm after washing, which creates handling challenges,” adds Hoffman. “And, anything with a thin-wall casting can be difficult to check for leaks, due to part expansion at higher pressures.”
“The bigger the part, the more temperature-sensitive things are,” notes Linda Moran, director of marketing at Sciemetric Instruments Inc. “Advanced temperature compensation is often required. One [of our] customers, for example, washes the casting right after it is machined, which brings the part temperature up 20 C above ambient. The leak tester needs to compensate for this.”
Leak rate specifications have become much tighter in recent years as automakers and suppliers address a wide variety of emission standards enacted by the U.S. Environmental Protection Agency and other organizations.
“Some components have stricter leak rates than others and are directly dependent on the fluid it contains or if hermeticity is a requirement,” says Michael McKay, general manager of the air leak test division of VIC Leak Detection. An example would be cooling components, such as an air-conditioning condenser. It would have a leak rate requirement based on the type of refrigerant gas used.
“Test pressure also impacts fixture design,” McKay points out. Large parts and high test pressure require large clamp forces to seal.”
Sealing is much more straightforward with a machined surface. Because cast parts feature unfinished, rougher surfaces, sealing can be quite challenging. “That is the trickier part of fixturing,” warns Moran.
McKay and other industry experts agree that engine blocks are the most challenging type of large casting to leak test. “Engine blocks are often more difficult than other large castings, such as aluminum wheels, because they have multiple cavities for oil, coolant and air,” he explains. “The cavities typically have different leak rates and test pressures.”
Transmission cases, transaxle cases and continuously variable transmission (CVT) cases also fall into this category,” adds Doug Dalgliesh, president of FlexControl Solutions LLC, which is the North American distributor for Yamaha Fine Technologies. “Sealing can be an issue with those castings. For example, engine blocks and transmission cases have half-round holes,90-degree vertical surfaces and many oil holes, which makes it difficult to seal the unit for helium leak testing.”
“What can make testing difficult are the number of openings that need to be plugged, or isolating a particular portion of a casting, such as the cooling channels on an engine block, for the test,” adds McLaren.
Other things to consider when testing large castings include the amount of time that’s involved. “It takes more time to run the test and more advanced techniques to get accurate readings,” says Sciemetric’s Moran. “To optimize what can be long cycle times and to manage the multiple in-line test stations typically required to maintain production volume, it is important to have as much process signature data as possible from the tests to maintain accuracy and reliability of the results.
One of the most difficult aspects of testing a cast part is the porosity of the metal. If a casting is found to leak during a vacuum chamber test, locating the source of the leak may be difficult. “Because seal life is always a concern, we use O-rings wherever possible,” says VIC’s McKay. “They’re inexpensive and effective.”
“With other test methods, porosity will show up as a leak because pressure or flow measurement devices will detect a change as gas seeps into the pores in the casting,” McLaren points out. “Helium leak testing detects gas that escapes from the part, so only true leaks are detected, reducing false positives.”
Leak Testing Methods
A variety of leak testing methods can be used on large automotive parts, including helium, air, mass flow and pressure decay. “Typically, large castings are tested by either helium or bubble test,” says ULVAC’s Goebel. “Manufacturers use bubble testing for less expensive parts used in all-terrain vehicles, recreational vehicles and tractors, due to the lower production volumes, as well as final product price point.
“Helium leak testing via either sniffer method or chambered automatic leak test systems are [commonly used for car parts],” adds Goebel. “[For applications] where production throughput is critical, such as aluminum wheels, which average . . . test times less than 8 seconds per part, helium is required.”
Goebel says hydrogen is not a viable tracer gas due to its inherent “stickiness,” which can eventually lead to elevated background levels, causing a false fail. “It can work in long test cycle times or in sniffer methods whereby inherently leaked gas can escape upwards [without] impacting a test or future tests,” he explains.
According to Agilent’s McLaren, helium “offers the best sensitivity, can be used in either vacuum or sniffing modes, and when used properly, is the most economical of all trace gasses.” Helium leak testing can be performed in several ways, such as sniffing, accumulation and spray probe.
“Because the specific test requirements, such as throughput, sensitivity, leak location and total leak rate, are so variable, any one of these methods may be used for testing castings,” McLaren points out. “And, with helium, only true leaks are detected and test equipment is unaffected by variations in porosity or temperature.”
Despite all those benefits, helium has some potential drawbacks. For instance, a helium shortage may be looming in the future. Some manufacturers with helium leak testing equipment are looking into testing with alternate gases, such as hydrogen and argon, which can also be expensive.
Flow-based methods are commonly used to test large automotive castings, because they’re less volume dependent. “Helium is mostly used only for testing aluminum wheels,” says InterTech’s Hoffman, who says dry air makes more sense. “Mass flow leak testing technology is ideally suited for [most other] large volume parts,” he claims. “It gives a direct reading and offers faster cycle times, which is critical for automotive applications.”
Transmission castings are often tested with air using a mass flow instrument. For instance, on a transmission cover, the casting cavity is typically tested to a limit of 10 standard cubic centimeters per minute (sccm), while the hub pathway and piston area are tested to a limit of 0.6 sccm.
Transmission cases, transaxle cases, CVT cases, covers, oil pans and DC-DC converter cases are usually tested by pressure decay, because [they contain] many oil holes,” says FlexControl’s Dalgliesh. “[This technique is also popular] due to the edge of connecting surfaces that have to be sealed completely if [the part] is tested using a helium leak tester. Also, leakage of oil holes is not very small.
“However, some transaxles have water cooling lines,” adds Dalgliesh. “In this case,helium leak testers can be used with precision threshold requirements. Cylinder blocks and cylinder heads are typically tested with a combination of pressure decay and water bubble methods.”
No matter what type of leak test technology they use, automakers and suppliers demand equipment that is fast and sensitive. The universal goal is to find the smallest possible leak in the shortest amount of time.
Flexibility is also important, so that the leak test station can handle different model types. “End users want more flexibility in testing techniques and configurations,” says Gene Grilli, director of global sales at Uson LP. His company recently unveiled a flexible tester called the Optima vT that supports a dozen different types of test methods.
“Engine blocks and transmission case castings can be done concurrently with eight independent test sensors on two channels,” Grilli points out. “Because pneumatic controls can be custom-configured to application requirements, it [can be] easily adapted to the large variety of part configurations in cast units such as engine blocks.
“One of the other features we have built into this new tester is a custom data collection [function] that allows the part to be tracked by bar code or serial number,” adds Grilli. “Tracking a history of the leak rates of subassemblies, and the complete assembled engine or transmission, makes data analysis easier [and allows engineers] to catch process and product problems earlier.”
All automotive manufacturers want more data management and traceability. “In leak testing, that means more information than straight pass-fail,” says Sciemetric’s Moran. “There are simply too many variables at work during the leak test to rely on limited data to keep the process under control. We see manufacturers struggling with trying to figure out how to reduce cycle times or explain variations between test stations or find the root cause of problems.”